finetune

# @package _global_

# specify here default configuration
# order of defaults determines the order in which configs override each other
defaults:
  - _self_
  - data: pytorch_dataset
  - model: supervised
  - callbacks: default
  - logger: null # set logger here or use command line (e.g. `python train.py logger=tensorboard`)
  - trainer: default
  - paths: default
  - extras: default
  - hydra: default

  # experiment configs allow for version control of specific hyperparameters
  # e.g. best hyperparameters for given model and datamodule
  - experiment: null

  # config for hyperparameter optimization
  - hparams_search: null

  # optional local config for machine/user specific settings
  # it's optional since it doesn't need to exist and is excluded from version control
  - optional local: default

  # debugging config (enable through command line, e.g. `python train.py debug=default)
  - debug: null

name: "finetune"
# task name, determines output directory path
task_name: null

test_devices: "1"

# tags to help you identify your experiments
# you can overwrite this in experiment configs
# overwrite from command line with `python train.py tags="[first_tag, second_tag]"`
tags: ["dev"]

# set False to skip model training
train: True

# evaluate on test set, using best model weights achieved during training
# lightning chooses best weights based on the metric specified in checkpoint callback
test: True

# simply provide path to previous train run to finetune from
pretrain_path: ???
pretrain_ckpt_path: ${pretrain_path}/checkpoints/best_model.ckpt
pretrain_yaml_path: ${pretrain_path}/hydra_config.yaml
best_config_path: ${pretrain_path}/best_config.json

ckpt_path: null

# seed for random number generators in pytorch, numpy and python.random
seed: null

# When tuning, setting this to 0 will desable ray memory monitor, which often crashes
ray_memory_monitor_refresh_ms: "0"